1. Top of page
  2. Abstract
  3. Introduction
  4. Glass Lasers: An Overview
  5. Current High-Power Glass Lasers
  6. Laser Glass Properties Important for High-Power Laser Applications
  7. Laser Glass Compositions
  8. Laser Glass Manufacturing
  9. Glasses to Meet Future HEHP, HAP, and PW Laser Needs
  10. Acknowledgments
  11. References

Advances in laser glass compositions and manufacturing have enabled a new class of high-energy/high-power (HEHP), petawatt (PW), and high average power (HAP) laser systems that are being used for fusion energy ignition demonstration, fundamental physics research, and materials processing, respectively. The requirements for these three laser systems are different, necessitating different glasses or groups of glasses. The manufacturing technology is now mature for melting, annealing, fabricating, and finishing of laser glasses for all three applications. The laser glass properties of major importance for HEHP, PW, and HAP applications are briefly reviewed and the compositions and properties of the most widely used commercial laser glasses are summarized. Proposed advances in these three laser systems will require new glasses and new melting methods, which are briefly discussed. The challenges presented by these laser systems will likely dominate the field of laser glass development over the next several decades.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Glass Lasers: An Overview
  5. Current High-Power Glass Lasers
  6. Laser Glass Properties Important for High-Power Laser Applications
  7. Laser Glass Compositions
  8. Laser Glass Manufacturing
  9. Glasses to Meet Future HEHP, HAP, and PW Laser Needs
  10. Acknowledgments
  11. References

This year (2011) marks the 50th anniversary of Snitzer's1 first glass laser. This discovery is chronicled in his own words in a remarkable personal interview published by the American Institute of Physics.2 Few are aware that Snitzer's great invention came on the heels of unjust accusations and treatment by the infamous 1950s U.S. House of Representatives Committee on Un-American Activities. Snitzer's work during that time proved that he is not only a great scientist but also a man of principle and tremendous courage.

Snitzer's discovery spawned many new inventions in the field of lasers, optical devices, and fiber communications. In this manuscript, we focus on just one of these areas: glass lasers for high-power applications and the laser glasses (gain media) that comprise the heart of these lasers.

We begin with a generic description of a laser and then explain the Nd-doped glass laser and its operation in conceptual terms. Our description is not meant to supplant the many outstanding texts on lasers.3–6 Instead the goal is to provide a common starting point for the readers of this journal who may be unfamiliar with certain aspects of laser design and operation. Interested readers should particularly consult Koechner3 and Koechner and Bass6; these sources focus on solid-state lasers with significant treatment of glass lasers.

Next, we briefly describe the three main types of high-power glass lasers that are currently operational. Also included is a description of the laser glasses used in each of the three systems. This is followed by a summary of key laser glass properties and composition space important for high-power applications. A short description of the glass manufacturing methods available to meet current and near-term requirements is also presented.

One is lead to the conclusion that after 50 years of glass laser development, the suite of glasses suitable for use in high-power laser systems is surprisingly small. The reasons for this are discussed.

A case is made that developing new glasses with much improved thermal–mechanical properties, while maintaining good laser and optical properties, will be the biggest challenge for meeting the needs of the next generation of high-power glass lasers. Examples of promising glasses and the required manufacturing methods are discussed. Competition from single-crystals and transparent polycrystalline ceramics offer a potential replacement for laser glasses, but face significant challenges in achieving the quality, size, and throughput that are the hallmarks of optical glass melting.

Glass Lasers: An Overview

  1. Top of page
  2. Abstract
  3. Introduction
  4. Glass Lasers: An Overview
  5. Current High-Power Glass Lasers
  6. Laser Glass Properties Important for High-Power Laser Applications
  7. Laser Glass Compositions
  8. Laser Glass Manufacturing
  9. Glasses to Meet Future HEHP, HAP, and PW Laser Needs
  10. Acknowledgments
  11. References

In its simplest configuration, a laser consists of a gain-media positioned inside a resonate cavity (Fig. 1a). The gain-media is the photon generator; it can be a solid, liquid, gas, or plasma. Energy is stored in the gain media by means of “pumping” using an external energy source. The pump energy induces transitions in atoms or molecules from lower to higher energy levels (rotational, vibrational, and/or electronic), and under proper conditions, one achieves the necessary population inversion between two energy levels required for laser action.


Figure 1.  Schematic representation of (a) laser oscillator and (b) oscillator plus amplifier.

Download figure to PowerPoint

The resonant cavity that contains the gain medium is comprised of a high reflectivity mirror on one end and a partially reflective mirror (“output coupler”), on the other. The output coupler functions as the name implies, it transmits a fraction of the light circulating in the resonator cavity into an output beam. If the output coupler operates in a steady-state mode (constant energy output), it is termed “continuous wave” laser operation. In contrast, some output couplers are designed to abruptly change from near 0% transmission to 100%. This configuration allows one to dump all or part of the energy circulating in the resonator into a single pulse; hence, the term pulse-mode operation. The high-power glass lasers that are the focus of this article operate in the pulsed-mode.

The system comprised of the resonator cavity, gain medium, and pump source is commonly called a laser “oscillator.” A variant of the simple oscillator design, of importance for the discussions here, is shown in Fig. 1b. Here, the output from the oscillator is directed to a second pumped gain medium that functions as an amplifier. There are many advantages to the oscillator/amplifier concept, the chief being the ability to generate high-energy/high-power laser outputs. For example it is straightforward to imagine laser designs using multiple amplifiers and/or multipass amplifier configurations to generate high-power output. Such design concepts form the basis for the high-energy/high-power (HEHP) glass lasers in operation today.

Glass lasers are a subset of solid-state lasers where the gain medium consists of rods or plates (slabs) of optical quality glass doped with a lasant ion.3,6 A common dopant ion, and the one used by Snitzer in his original laser, is Nd3+. Typically, Nd3+-doped laser glass plates or rods are installed in a flashlamp-pumped cavity as shown schematically in Fig. 2. When the laser is “fired,” a capacitor bank sends an electrical discharge through a xenon-containing flashlamp, which in turn produces an intense pulse of white-light with a spectral distribution that overlaps the absorption bands of the Nd3+.


Figure 2.  Schematic representation of a flashlamp-pumped amplifier containing rectangular slabs of laser-glass gain media.

Download figure to PowerPoint

The pumping and amplification process by the Nd-laser glass can be described in terms of either an engineering or atomistic representation as shown in Fig. 3. In the simpler “engineering” representation (Fig. 3a), the laser glass stores optical energy delivered by the flashlamps. The stored energy in the glass is characterized by a gain coefficient, g. As a weak input pulse (i.e., small signal) propagates through the glass, the pulse energy increases exponentially with the distance, L (cm), as given by5:

  • image(1)

where Iin and Iout are the pulse input and output intensity (W/m2), respectively. The small-signal gain, Go, of the system is defined as the ratio of the output to input intensity:

  • image(2)

Figure 3.  (a) Engineering and (b) atomistic representations of flashlamp pumping, Nd excitation, relaxation, and stimulated emission (amplification) in Nd3+-doped laser glass.

Download figure to PowerPoint

It is important to recognize the distinction between the gain coefficient, which is a material property, and the gain, which is a system performance measure. The unfortunate choice of wording often leads to confusion in the use of these two terms.

In the more fundamental “atomistic” representation (Fig. 3b), the flashlamp light excites (“pumps”) the Nd3+ 4f electrons from the ground state (4I9/2) to a manifold of higher energy states. The electrons rapidly relax to the metastable 4F3/2 state by a series of nonradiative energy transitions to glass phonon modes (heat). The nonradiative loss rate is slow between the metastable 4F3/2 level and the lower lying 4I states because of the large energy gap between them. This leads to an accumulation of electrons at this level and generates the population inversion necessary for laser action. A weak laser pulse, with a wavelength matched to the energy between the 4F3/2 and 4I11/2 states, stimulates the transition between these two states and coherently adds energy to (amplifies) the transmitted pulse. The electrons in the terminal laser level (4I11/2) rapidly relax to the ground state. Note that Nd3+ is an example of a “four-level” laser system3,7: (1) ground state4I9/2 (2) the manifold of pump bands (3) upper laser level—4F3/2, and (4) terminal laser level—4I11/2.

The atomistic representation allows one to express the gain coefficient (g) in terms of more fundamental properties of the Nd3+ glass system:

  • image(3)

where σ(λ) is the Nd3+ emission cross section (cm2) at wavelength λ, inline imageis the Nd3+ inversion density at the 4F3/2 state (1/cm3), and α is the absorption coefficient (/cm) at the laser wavelength. The emission cross section is a measureable quantity and physically represents the effective interaction area of Nd3+ in the 4F3/2 state with a propagating photon of wavelength (λ). Consequently, the greater the cross section the more likely the interaction and the greater the gain achieved.

Note from Eq. (3) that the gain coefficient is comprised of two terms. The first term (inline image) describes the amplification due to the stimulated emission from the upper laser level, whereas the second defines the energy loss due to absorption (α) by impurities in the glass. Therefore, to achieve a large gain requires a glass composition for which the Nd3+ emission cross section is maximized, combined with efficient optical pumping to achieve a large population inversion (inline image). In addition, the glass must be manufactured with a high purity to minimize transmission losses due to absorbing impurity ions.

It is instructive to compare the magnitude of various terms in Eq. (3) for a typical flashlamp-pumped amplifier; for example, consider a high-power laser amplifier used on a fusion energy research laser (these lasers are described in the next section). A typical amplifier small-signal gain coefficient has a value of about 0.05/cm. In other words, the weak laser pulse is amplified exponentially by 5% for every cm of glass it propagates through. The absorption loss (α) of the laser glass is measured during manufacture and is typically <0.001/cm; thus, the loss is <1/50 of the gain.

The measured Nd3+ emission cross section for the glasses used in these lasers is about 4 × 10−20 cm28 and can be used to estimate the excited-state density in the metastable state from the gain coefficient (i.e., inline image). inline image is computed to be approximately 1.3 × 1018/cm3 based on a value of 0.05/cm for g. The excited-state density, inline image, has greater physical meaning when expressed in terms of the “stored energy” in the glass. Recall that each electron in the metastable state can potentially produce a stimulated laser photon of energy, hν (1.885 × 10−19 J), and the product inline image is the optical energy stored in the glass that could be extracted during a laser shot. The stored energy in the laser glass is about 250 J/L for this example. The amount of the stored energy that is extracted by the incident laser pulse depends on a number of factors (laser design, pulse intensity, pulse length, etc.) and can range from less than a few percent to as much as 80%.

The Nd-doped laser glasses used in high-power systems are predominantly phosphate based with near metaphosphate compositions. The reasons for this choice are described in more detail later, but in brief, Nd-phosphate glasses present the following main advantages:

  • Large stored energy.
  • Efficient energy extraction.
  • Resistance to laser-induced damage.
  • Mature manufacturing technology.

The choice of which specific phosphate glass to use depends on the type of glass laser. In the next section, the three main classes of high-power glass lasers and their primary applications are described. This provides the launching point for the subsequent discussion of laser glass properties and compositions.

Current High-Power Glass Lasers

  1. Top of page
  2. Abstract
  3. Introduction
  4. Glass Lasers: An Overview
  5. Current High-Power Glass Lasers
  6. Laser Glass Properties Important for High-Power Laser Applications
  7. Laser Glass Compositions
  8. Laser Glass Manufacturing
  9. Glasses to Meet Future HEHP, HAP, and PW Laser Needs
  10. Acknowledgments
  11. References

In general, current high-power glass lasers can be divided into three broad classes as given in Table I. HEHP and PW laser systems are essentially single-shot devices designed to be fired and then allowed to cool several hours before refiring. In contrast, the high average power (HAP) lasers are designed to operate continuously at repetition rates, typically in the range of ∼1–10 Hz.

Table I.   The Three Main Categories of High-Power Glass Lasers in Operation Today and Typical Operating Characteristics
CategoryTypical laser output characteristics
Energy (J)Pulse length (s)Peak power (W)Repetition rate (Hz)
High-energy/high-power (HEHP)105–10610−9–10−81013–101510−4–10−5
Petawatt (PW)102–10310−14–10−13>101510−4–10−5
High average power (HAP)10–10210−9–10−81010–10111–10

HEHP Lasers for Inertial Fusion Research

HEHP glass lasers are predominantly used for Inertial Confinement Fusion (ICF) research. 9,10 There are currently multikilojoule HEHP glass laser systems in operation around the world (Table II); the major centers are in Japan,11 China,12,13 Russia,14 France,15 and the United States. Within the United States, there are three major facilities, all sponsored by the U.S. Department of Energy: the Laboratory for Laser Energetics at the University of Rochester, LLE,16,17 the Beamlet laser at Sandia National Laboratory,18,19 and the National Ignition Facility (NIF) at Lawrence Livermore National Laboratory (LLNL).20,21

Table II.   Examples of Multikilojoule HEHP Glass Laser Systems in Operation, Under Construction, or Proposed
Laser facility and statusLocationPhosphate glass typeLaser output at 1.05 μm (kJ)Laser output at 0.35 μm (kJ)
  • *

    At 532 nm.

  • HEHP, high-energy/high-power; NIF, National Ignition Facility; LLNL, Lawrence Livermore National Laboratory.

I. Operational
 OmegaUniversity of RochesterLHG-85430
 Omega-EPUniversity of RochesterLHG-82.7NA
 Gekko-XIIOsaka, JapanLHG-81210*
 BeamletLLNL/Sandia NLLG-7501710
 LFEX (PW)Osaka, JapanLHG-81NA
 Shenguang (SG)-IIUChinaN-313018
II. Under construction
 FireXOsaka, JapanLHG-850?
 Shenguang (SG)-III-TILChinaN-313015–20
 Shenguang (SG)-IIIChinaN-31300150–200
III. Proposed
 ISKRA-6RussiaKGSS-0180/35 grade600?
 Shenguang (SG)-IVChinaUndetermined30001500–2000

All major HEHP glass laser systems use Nd-doped phosphate glasses; Table II provides the commercial code designation and Table III a compilation of key properties for those in most common use. The majority of these glasses originate from two commercial suppliers: Schott22 and Hoya.23 Smaller quantities and sizes have also been provided by Kigre Inc.24 for certain lasers in the United States. The glasses for the lasers in China and Russia are manufactured within each country under the code designations N31 and KGSS-0180/35-grade, respectively (Table III). Brief descriptions of the melting technology and properties of the N31 glass used in China have been published recently,25 and, as stated by the authors, mimic closely the glasses and manufacturing methods developed by Schott and Hoya with LLNL.8,26–31

Table III.   Properties of Most Commonly Used Commercial HEHP Glasses
Glass manufacturerHoya23Schott22Kigre24
Glass propertiesSymbolLHG-80LHG-8LHG-5LG-770LG-750LG-760Q88Q98
  • *

    Nd self-quenching increases quadratically as [Nd/Q]2; see text.

  • Temperature not specified.

  • 20–300°C unless otherwise stated.

  • §


  • HEHP, high-energy/high-power.

 Refractive index
  @ 587.6 nmnd1.54291.52961.54101.50861.52601.51901.54491.555
  @ 1053 nmnl1.53291.52011.53081.49961.51601.50801.53631.546
 Nonlinear refractive index
  (10−13 esu)n21.
  (10−20 m2/W)γ3.363.133.452.852.982.883.373.55
 Abbe numberν64.766.563.568.468.269.264.863.6
 Temperature coefficient  refractive index (10−6/K)dn/dT−3.8−5.3−0.4−4.7−5.1−6.8−0.5−4.5
 Temp-coefficient optical path (10−6/K)δ1.−0.42.70
 Emission cross section (10−20 cm2)σem4.
 Saturation fluence (J/cm2)Fsat4.
 Radiative  lifetime (zero-Nd) (μs)τo337365320372383330326308
 Judd-Ofelt radiative  lifetime (μs)τr327351320349367320326
 Emission band width (nm)Δλeff23.926.526.125.425.323.521.925.5
 Concentration quenching  factor (cm−3)*Q10. 
 Fluorescence peak (nm)λL10541053105410531053105410541053
 Thermal conductivity (W/m K)  (298 K)k0.630.580.770.570.490.570.840.82
 Thermal  diffusivity (10−7 m2/s)DT3.
 Specific heat (J/g K)Cp0.630.750.710.770.720.750.810.80
 Coefficient of thermal  expansion (10−7 K−1)αe13012798134130150104§99§
 Glass transition temperature (°C)Tg402485455461450350367450
 Density (g/cm3)ρ2.922.832.682.592.832.602.713.10
 Poisson's ratioμ0.
 Fracture toughness (MPa m0.5)KIC0.460.510.420.480.450.47
 Hardness (GPa)H3.353.434.23.582.853.18
 Young's modulus (GPa)E50.050.167.747.350.153.769.870.7
 Thermal shock  resistance (W/m1/2)Rs0.330.340.370.320.250.24

The glass property compilation in Table III represents the most recently reported values, to the best of our knowledge. Most come from current company marketing data sheets, recent measurements in our own laboratories or from recent literature sources as so noted. Values for some properties may differ slightly (generally <10%) from earlier compilations due to measurement variations between laboratories, improved or new measurement methods, and/or small changes in compositions; for example, the Nd-doping level can cause small variations in many properties. Also, some laboratories report measurements at slightly different temperatures or over different temperature ranges (e.g., thermal expansion coefficient); these are noted in Table III, when reported. In any event, the differences are small and should have a minor impact for most applications; if needed more information can be obtained by contacting the glass vendor. (This same comment applies to the property compilation of HAP glasses (Table IV) as discussed in the next section.)

Table IV.   Properties of Commercial HAP Glasses and One Developmental Glass by Fujimoto et al.78 (Nd-SG)
Glass manufactureKigre24Schott22Hoya23Development78
Glass propertiesSymbolQ89-NdAPG-1APG-2HAP-4HAP-3Nd-SG
  • *

    Computed by authors using J. O. treatment of reported spectral data in Fujimoto et al.78

  • Nd self-quenching increases linearly as [Nd/Q]; see text.

  • 20–300°C.

  • §Value at 25°C.

  • HAP, high average power.

 Refractive index
  @ 587.6 nmnd1.5591.53701.51271.54331.52981.4584
  @ 1053 nmnl 1.52601.50321.53311.52001.4496
 Nonlinear refractive index
  (10−13 esu)n2
  (10−20 m2/W)γ 3.112.983.383.052.53
 Abbe numberν63.667.766.964.667.767.9
 Temperatue coefficient refractive  index (10−6/K)dn/dT1.
 Temperatue coefficient optical  path (10−6/K)δ5.
 Emission cross section  (10−20 cm2)σem3.
 Saturation fluence (J/cm2)Fsat5.
 Radiative lifetime (zero-Nd) (μs)τo350385464350380376(?)
 Judd–Ofelt radiative lifetime (μs)τr361456372512*
 Emission band width (nm)Δλeff21.227.831.527.027.951.7
 Concentration quenching factor  (cm−3)Q16.710.6
 Fluorescence peak (nm)λL10541053.91054.610541052.51062
 Thermal conductivity (W/m K)  (90°C)k0.820.830.841.02§0.80§1.38
 Thermal diffusivity (10−7 m2/s)  (90°C)Dt3.
 Specific heat (J/g K)Cp0.840.770.710.74
 Coefficient of thermal expansion  (10−7/K)αe8899.662.68583∼1
 Glass transition temperature (°C)Tg440450549486541
 Density (g/cm3)ρ3.142.632.562.702.662.20
 Poisson's ratioμ0.
 Fracture toughness (MPa m0.5)KIC0.600.640.830.480.80
 Hardness (GPa)H3.094.7
 Young's modulus (GPa)E7064697572.7
 Thermal shock resistance  (W/m1/2)Rs0.541.031.100.47∼20 estimated

One example of a mega-joule scale HEHP laser system is the NIF that began full-scale operation in 2009 at the LLNL.20,21 The name of the facility is derived from its purpose: to achieve controlled thermonuclear (fusion) ignition in a laboratory setting.9,10 The NIF is the largest laser, as well as largest optical system, ever constructed32 and is capable of irradiating mm-size targets contained in a 10-m-diameter target chamber with energies up to 1.8 MJ at 351 nm and peak powers of 5.0 × 1014 W (500 TW).21,33 The NIF contains >7500 large optics of 40 cm or greater transverse size including laser amplifier glass slabs, lenses, mirrors, polarizers, and crystals and an additional 26,000 smaller optical components (<20 cm). The total area of precision optical surfaces in NIF is nearly 4000 m2. The building that houses the laser, beam propagation optics, and target chamber covers an area >104 m2.

The NIF utilizes 192 identical laser beam-lines to achieve its mega-joule output energies.20 The main glass laser amplifiers are grouped in two laser bays, each of which contains half (96) of the laser beamlines (Fig. 4). Each beam-line contains 16 precision-polished, Nd-doped, phosphate glass plates (3072 for the full system) and each plate measures 81 cm × 46 cm × 4 cm. The laser glass plates are installed in a series of cassettes (Fig. 5), which are then installed in flashlamp-pumped amplifiers. When the laser is “fired,” a capacitor bank sends a 0.36 ms electrical discharge (60 kV and 25 kA) through 7680 xenon-containing flashlamps (each ∼2.5 m long). The flashlamps generate an intense pulse of white-light (∼30 kJ/lamp) with a spectral distribution that overlaps the absorption bands of the Nd3+ in the laser glass (see Fig. 3b). The NIF uses two commercially available laser glasses in the main amplifiers: Schott's LG-770 and Hoya's LHG-8 (Table III). The glasses have an Nd-doping density of 4.2 × 1020Nd/cm3.


Figure 4.  Photograph of one of the two laser bays of the NIF; note the workers, in yellow hard hats, in the lower left corner for scale. Each bay accommodates half of the 192 individual laser beams that comprise the NIF laser system. The ∼40 cm × 40 cm aperture laser beams are transported in large diameter pipes clearly visible in the photo. The main laser amplifiers are assembled in the laser bay at given positions along the beamlines.

Download figure to PowerPoint


Figure 5.  (a)Precision laser glass plates are assembled in (b) cassettes for installation on the NIF. A total of 768 such cassettes are used in the NIF amplifiers.

Download figure to PowerPoint

HAP Lasers for Material Processing

HAP glass laser systems are predominantly used for commercial materials processing, particularly laser peening (LP) of metals. LP is a surface treatment process developed to improve fatigue performance and the strength of high-value metal parts, particularly those used in aerospace applications. Extensive literature exists on the LP process and two recent sources contain a thorough compilation of much of this work.34,35

To our knowledge, only two companies provide this capability in a commercial setting: Metal Improvement Company Inc.36 and LSP Technologies Inc.37 Each has proprietary HAP laser systems and peening technology. Both companies use commercial laser glasses specifically formulated for high-power applications (Table IV). Metal Improvement Inc. operates in the United States and Europe and also fields self-contained mobile laser systems on a semitrailer that can be transported to the customer's site.36 The LP beam is then “piped” to the work piece; in principle, there is no size limit to the work piece, and if so required, could be a full-size aircraft, a ship, or other large structure.

The LP process is schematically shown in Fig. 6. In brief, the pulsed output from an HAP laser is used to irradiate either the bare metal surface or a thin sacrificial ablation layer that has been applied to the surface. The laser interaction with the part surface causes it to rapidly heat, vaporize, and partially ionize, generating a plasma with temperature ∼104 K (∼1 eV). The rapid vaporization and expansion of the ablated material are inertially confined by a thin layer of water that flows laminarly over the LP surface and serves as a tamper that intensifies the pulse pressure. The ablation pressure scales as the square-root of the irradiance. Typically, an irradiance of 10 GW/cm2 (at ∼15–20 ns pulse width) generates a peak pressure of ∼2.5 GPa (25 kbar). The pressure pulse launches a shock wave that propagates through the metal. The plastic deformation of the metal by the shock wave produces a permanent compressive residual stress that penetrates to a depth of between 1 and 8 mm, depending on the material and the processing conditions. Multiple firings of the laser in a predefined spatial pattern imparts a layer of residual compressive stress over the desired area of the part. The compressive stress layer creates a barrier to crack initiation and growth, which consequently enhances the fatigue life and can provide resistance to stress corrosion cracking and fretting.34,35


Figure 6.  Schematic representation of the laser shock peening process showing (a) the laser pulse incident on the work piece, and (b) the resulting high-pressure ablation plasma and subsequent pressure (shock) that propagates through the material.

Download figure to PowerPoint

An example of a HAP glass laser used for laser shock peening has been described by Dane and colleagues.38,39 The laser uses a “zig-zag” amplifier design (Fig. 7), referring to the fact the extraction pulse propagates within the glass in a zig-zag pattern controlled by total-internal-reflection at the slab/water interface. The slab is positioned in the center of the amplifier assembly and has water cooling channels along the sides formed by the slab face and a clear glass window. Flash-lamps located along the slab faces pump the laser glass through the window. The glass is configured in a thin slab shape to enhance the conductive heat removal from the slab center to the cooling water. Dane et al.39 report slab dimensions of ∼1 cm × 14 cm × 40 cm with nominal Nd-doping levels of ∼3 × 1020/cm3.


Figure 7.  Highly schematic representation of a flashlamp pumped zig-zag amplifier commonly used on HAP laser systems.

Download figure to PowerPoint

Petawatt (PW) Glass Lasers for Basic Science and Advanced Fusion Research

PW glass laser development has shown tremendous growth since the demonstration by Perry and colleagues40,41 of a 1.25 PW (1.25 × 1015 W) Nd-glass laser. This growth has been driven largely by the realization that the so-called “table top” PW lasers can been built with modest funds, thus making these systems available to university researchers. PW powers provide access to a new regime of laser–matter interactions of interest for fundamental physics studies.42,43 In addition, Tabak et al.44 proposed a “fast ignition” scheme for achieving hot-spot fusion ignition with PW scale lasers that has spurred significant interest in the ICF research community.

Forty-five centers for PW laser research are listed in a 2010 compilation provided by the International Committee on Ultra-high Intensity Lasers45 and include sites in the United States, Canada, England, France, Russia, Italy, Germany, Japan, China, India, and South Korea. In general, the PW systems fall into two major categories: (i) large-scale, high-energy systems that are constructed as add-ons to existing multikilojoule fusion-research lasers11,15,17,21 and (ii) smaller scale, lower energy “table-top” lasers built mainly at universities.42

To generate PW powers requires simultaneously achieving two often conflicting operating conditions: short pulse lengths (femto to pico-seconds) and modest to high energies (>100 J). Either condition by itself is readily accomplished with today's laser technologies. However, to achieve both simultaneously requires amplification at short pulse lengths, which in turn can lead to unacceptable laser-induced damage to the amplifier gain medium. The damage is a consequence of the intensity dependence of the refractive index of the gain media leading to self-focusing as discussed more in “Laser glass properties important for high-power laser applications.”

The approach used to conquer this problem is to first generate the required short pulse, then temporally “stretch” it to a longer pulse length (i.e., lower intensity) for amplification and finally recompress it back to the initial pulse length (Fig. 8). Many PW laser researchers use laser-pumped titanium-doped sapphire (Ti3+:Al2O3) to generate the initial short pulse. Ti-sapphire offers high-gain over a broad bandwidth, which is essential to short pulse generation. Chirped-pulse amplification (CPA) is then used to stretch, amplify, and recompress the pulse. CPA was first successfully demonstrated for solid-state lasers in the late 1980s at the University of Rochester46 and revolutionized PW laser research.


Figure 8.  Schematic representation of a typical Petawatt CPA design showing the short pulse temporal expansion, amplification, and compression.

Download figure to PowerPoint

It was recognized early on that an all-Ti:sapphire laser is inadequate for achieving the necessary PW energy and power output. Instead, a hybrid system was developed using a laser-pumped Ti:sapphire oscillator plus preamplifier for short pulse generation followed by flashlamp pumped Nd-doped glass power-amplifiers in the long pulse section. The short pulse output is typically “stretched” in time by a factor of ∼104 from the few hundred femtosecond to the nanosecond range. The resulting weak nanosecond pulse is then amplifed using conventional HEHP glass power-amplifer technology to energies in the range of ∼102–103 J. Matched sets of grating-pairs are used in the stretcher and compressor sections. In fact, a key optical technology for the success of CPA has been the development of advanced gratings.47 Futher details of the design of PW CPA systems are beyond the scope of this review; instead, the reader is referred to41,42,48 and work cited there in.

Laser glasses used in PW systems must meet two somewhat conflicting requirements: broad emission bandwidth and low saturation fluence. PW lasers require gain media with broad spectral bandwidths (Δν) to accommodate the short pulse widths.40,48 This limitation is generally expressed by the pulse-length (tp) times bandwidth (Δν) product (FWHM), which for a chirped gaussian-shape pulse is5,40:

  • image(4)

The exact value of the product varies somewhat depending on the pulse shape.5 Nevertheless, the issue remains the same: laser glasses used to amplify very short temporal pulses must have broad gain (emission) bandwidths.

The glass should also have a low saturation fluence (Fsat) to effectively amplify the pulse (i.e., extract energy from the glass) without laser damage or wavefront distortion. The saturation fluence is given by:

  • image(5)

where hν is the energy of the laser photon (J) and σ is the emission cross section of the lasant ion (cm2), in this case Nd3+. Physically, a saturation fluence of 1 represents a progating energy density (fluence, J/cm2) equivalent to one laser photon per unit-area equal to the Nd3+ emission cross section. To achieve a low saturation fluence requires a high-emission cross section. The conflict in desired amplifier glass properties arises because the lasant ion cross section and emission bandwidth are inversely related (see “Emission cross section and quantum yield”). Consequently, a low saturation fluence represents a small bandwidth and vice versa.

Some PW researchers have turned to using a combination of glasses to achieve the required performance. Hays et al.48 recently proposed a multi-PW to exawatt laser based on a clever combination of a suite of Nd:doped glasses with varying cross sections, bandwidths, and peak emission wavelengths. The glasses were selected from an extensive compilation of laser, optical, and physical property data on nearly 250 Nd-doped glass compositions prepared and characterized during the 1970s and 1980s. Table V lists several laser glasses in use on PW systems, including the ones proposed by Hays and colleagues.

Table V.   Properties of Commercial Laser Glasses in Use on Some PW Systems and Developmental Glasses for Potential Future Use
Glass codeCross section σ (10−20 cm2)Bandwidth, Δλeff (nm)Radiative lifetimeτrad (μs)Nonlinear refractive index n2 (10−13 esu)Saturation fluence Fsat (J/cm2)Glass type
  • *


  • Kigre.24

  • Fujimoto et al.78

  • §

    LLNL glass catalog.90

I. Commercial
II. Developmental

Laser Glass Properties Important for High-Power Laser Applications

  1. Top of page
  2. Abstract
  3. Introduction
  4. Glass Lasers: An Overview
  5. Current High-Power Glass Lasers
  6. Laser Glass Properties Important for High-Power Laser Applications
  7. Laser Glass Compositions
  8. Laser Glass Manufacturing
  9. Glasses to Meet Future HEHP, HAP, and PW Laser Needs
  10. Acknowledgments
  11. References

In the previous section, we listed the key properties of laser glasses used on today's HEHP, HAP, and PW systems in Tables III–V, respectively. However, little explanation was given as to why these properties are important; this is the purpose of this section. Our treatment, however, is by no means exhaustive. Typically, >20 individual glass properties must be optimized or controlled to give the desired laser glass performance and this does not include many of the properties important for manufacturing. Here, we examine just a few key properties. Several sources and the reference listed there-in provide further details.8,49–52

Emission Cross Section and Quantum Yield

It can be argued that the two most important laser glass properties for high-power applications are the emission cross section and quantum yield. These two properties largely control the energy storage, gain coefficient, and extraction efficiency of the laser glass. For high-power applications, the general goal is to store and extract as much laser energy as possible.

The Nd3+ cross section is determined from the analysis of the absorption and emission spectra (Fig. 9). The emission spectrum corresponds to the 4F3/2 to 4I11/2 transition and usually peaks near 1.05–1.06 μm; this is the dominant Nd3+ transition and the one of greatest interest for laser applications. The emission band is characterized in terms of the peak emission wavelength (λp) and the effective bandwidth, Δλeff:

  • image(6)

Figure 9.  (a) Nd3+ absorption spectrum, (b) emission spectrum, and (c) emission transient for a typical Nd3+-phosphate glass. The emission transient is used to determine the emission lifetime, τmeas. Nonradiative energy losses reduce the lifetime and the quantum efficiency as discussed in the text and in Fig 10.

Download figure to PowerPoint

There are another three emission bands associated with transitions from the 4F3/2 to the other 4IJ states (i.e., 4I7/2,9/2,13/2), but they are considerably weaker. The relative strength of each emission band is expressed in terms of the branching ratio, BJ, which is defined as the fraction of the total fluorescence that terminates at a particular 4IJ state (J=7/2, 9/2, 11/2, or 15/2) and ΣBJ=1.

The absorption spectrum is used to determine the branching ratio and radiative lifetime, τrad, by an analysis method known as the Judd–Ofelt (J–O) treatment.53,54 In a classic paper, Krupke55 provided glass chemists with a road map for use of the J–O treatment to assess key laser glass properties from straight-forward spectroscopic measurements on samples from small-scale test melts. The J–O method is widely used and a number of texts describe the treatment in detail.7

The quantities determined from the spectral analysis and the J–O treatment (i.e., B11/2, λp, τrad, and Δλeff) are used to calculate the emission cross section using the Einstein relation5,7

  • image(7)

where n is the refractive index at λp and c is light speed (m/s). Cross sections for several commercial Nd-phosphate laser glasses are listed in Tables III–V and usually range from 3 to 4 × 10−20 cm2.

The quantum efficiency (ɛeff), sometimes called the quantum yield, represents the fraction of Nd3+ in the upper laser level (4F3/2) that relaxes via radiant emission (fluorescence) and is commonly defined as7,8:

  • image(8)

where τmeas is the measured emission lifetime. In practice, the quantum efficiency is never unity due to various nonradiative loss mechanisms that divert some fraction of the available stored energy to heat and measurably shorten the emission lifetime. These nonradiative losses are affected by the intrinsic properties of the laser glass as well as the purity with which the glass is manufactured.56,57Figure 10 presents a schematic representation of the most important Nd3+ nonradiative loss mechanisms in phosphate glasses.8,56 Impurities with absorption bands that overlap any of the four Nd3+ transitions from the 4F3/2 state to the lower lying 4I manifold (i.e., absorption in the 800–2000 nm range) are particularly troublesome. For this reason, OH and certain transition metal ion impurities (Cu2+, Fe2+) are of most concern. The use of very-high-purity raw materials and melter refractories is required to remove the threat of transition metal impurities. Similarly, specific processing steps are added to eliminate OH from the glass melts26,30; this dehydration step is one of the more difficult aspects of melting phosphate laser glasses because they tend to be hydroscopic.


Figure 10.  Schematic representation of major mechanisms for nonradiative energy loss from the 4F3/2 state (i.e., upper laser level): (a) Nd-to-Nd energy transfer (self-quenching), (b) energy transfer to OH vibrational modes, and (c) transition metal vibronic excitations.

Download figure to PowerPoint

Another important nonradiative loss mechanism is Nd-concentration quenching56 (Fig. 10). The mechanism entails energy transfer between two Nd3+ ions, one in the 4F3/2 excited state and the other in the 4I9/2 ground state. The exchange converts both to the 4I15/2 level from which they nonradiatively decay to the ground state. This nonradiative loss mechanism usually dominates at high Nd concentrations and is characterized by Q, the quenching factor. Q is defined as the concentration (ion/cm3) that reduces the lifetime by one half. Q is usually determined from emission lifetime measurements on a series of samples with different doping densities. The higher the Q value, the less sensitive (better) the laser glass is to Nd self-quenching. In theory, the degree of quenching increases as the square of the Nd-doping level because the Nd series serves as both the donor and acceptor for energy transfer. Measurements show HEHP glasses follow the expected quadratic dependence. However, HAP glasses often exhibit near linear behavior. Thus, the units for Q as listed in Tables III and IV reflect the specific functional dependence.

Nonlinear Refractive Index

The performance of HEHP and PW lasers can be limited by nonlinear propagation effects, particularly at high irradiance. Locally, the refractive index increases in the presence of an intense laser beam as given by3,5:

  • image(9)

where γ is the nonlinear refractive index coefficient (m2/W) and I is the laser irradiance (W/m2). This can lead to self-focusing effects that can cause damage to the downstream laser glass or other optics. Self-focusing is generally divided into two types (Fig. 11): (1) whole beam self-focusing that is a natural consequence of the Gaussian shape of many small-aperture laser beams and (2) localized self-focusing associated with spatial or temporal noise on the beam. For the case of whole-beam self-focusing, the higher intensities of the central regions of the beam induce a radial index gradient (lensing effect) causing the whole beam to focus down on itself. At some point, the irradiance exceeds the breakdown threshold of the material producing a damage spot in the optical material.


Figure 11.  Illustration of (a) whole beam and (b) localized self-focusing during propagation of a high-intensity laser beam through an optical material of refractive index n(I).

Download figure to PowerPoint

Localized self-focusing is similar in that high-intensity noise spikes on a large-aperture beam collapse (focus) and generate associated damage on optics along the beam path. For example, damage commonly occurs on coated optics (e.g., mirrors and polarizers) and in the bulk glass of lenses and amplifier slabs along the beam propagation path.

The threat of nonlinear noise growth is often expressed in term of the so-called “B-integral”:

  • image(10)

where B (radians) is the cumulative nonlinear phase retardation over the optical path length.5,58 Intensity ripples (noise) that occur at certain spatial frequencies grow exponentially with B:

  • image(11)

The greater the value of B, the greater the threat the beam will “break-up” into filaments as illustrated in Fig. 11. In fact the use of the letter “B” to represent this term is an abbreviation for “break-up.”58 Past experience has shown that B needs to be less than about 2 radians to avoid unacceptable noise growth.59,60

Direct measurement of γ is difficult, so empirical correlations have been developed. The expression developed by Boling et al.61 many years ago accurately predicts γ from the refractive index (nd) and the Abbe number (ν) of the glass:

  • image(12)

where K=2.8 × 10−10 m2/W and is an empirically determined constant.

Note that many researchers express Eq. (9) in terms of the nonlinear refractive index, n2 (in esu) rather than the more engineering friendly nonlinear refractive index coefficient, γ (m2/W); for linear polarized light, they are interrelated by n2=γ (nc/40π), where n is the refractive index at the wavelength of interest.

Thermal–Optical Properties

The laser glass must have high optical homogeneity to achieve the beam quality necessary to propagate and focus the laser output beam. Thermal variations in the laser glass can produce optical distortion by changing the optical path length. It is generally desirable to use a glass for which the temperature coefficient of the optical path length is zero; these are referred to as athermal glasses. The change in the optical path length, ΔOL, resulting from a temperature variation, ΔT, over a length, L, is given by:

  • image(13)

where δ is the temperature coefficient of the optical path length:

  • image(14)

and (dn/dT) is the temperature change in refractive index relative to air and αe is the coefficient of linear thermal expansion. An ideal athermal glass has δ=0 implying, from equation,14−dn/dT=(n−1)αe. Therefore, a good athermal laser glass must have a negative dn/dT value about half of the value of the coefficient of linear thermal expansion. Phosphate glasses are one of the few glass types that can be formulated to meet this condition.

Thermal–Mechanical Properties

Today's HEHP and PW laser systems tend to be essentially “single shot” devices with a few hours between shots; as a result, the glass physical properties tend to be of less importance. In contrast, HAP lasers operate at substantial thermal loading associated with rep-rated operation. For this reason, good thermal–mechanical properties such as a high fracture toughness, high thermal conductivity, and low thermal expansion are important. These properties are often collectively characterized in the well–known “thermal shock resistance,”RS (W/m1/2)62:

  • image(15)

where E is the Young's modulus (GPa), k the thermal conductivity (W/m K), KIC the fracture toughness (MPa m1/2), μ the Poisson's ratio, and αe the coefficient of linear thermal expansion (K−1). The thermal shock resistance equates directly to the maximum thermal load that a surface-cooled glass slab can tolerate before catastrophic fracture.62,63 Some versions of Eq. (15) include effects of surface flaws from finishing the glass; such flaws can seed fracture in externally cooled HAP glass plates.

Chemical Durability to “Weathering”

The resistance of glass to aqueous corrosion (“weathering”) is important for all high-power laser applications. Glass durability is generally characterized by measuring the solubility of the glass in water under some well-defined condition.

For the case of HEHP and PW glasses, maximizing the glass composition to achieve high durability eases the problems encountered with finishing, cleaning, storing, and handling the laser glass. The less durable glasses are more easily stained and fogged and can lead to optical damage or transmission losses or both. This in turn requires costly refinishing or the installation of expensive environmental controls within the laser system.

Glass durability requirements for HAP applications are much more demanding. HAP lasers are generally cooled with 100% deionized water or a mixture of deionized water and ethylene glycol and the amplifiers are expected to operate for years without maintenance or the need for slab replacement. Further complicating the durability issue is the threat of subcritical crack growth from small surface flaws generated during optical finishing or from laser-induced damage. Such flaws can grow to large fractures in the presence of thermally induced tensile stresses at the glass surface. As a consequence, HAP glasses are generally much more water durable than are HEHP compositions.

Laser Damage Resistance

The peak fluence in the laser glass of today's high-power lasers can approach 5–20 J/cm2 with a peak irradiance of up to 5.0 GW/cm2. To avoid optical damage, the laser glass surfaces must be precision polished and the bulk glass must be free of defects, specifically any microscopic inclusions that enter during processing. In most cases, the inclusions originate from the refractory wall of the glass melter. HAP lasers face the added threat of crack initiation and growth at inclusion damage sites in the presence of thermally induced tensile stress. In such cases, the glass slab can catastrophically fracture.62

The most common inclusions are ceramic or metallic particles from the liners used in the melting system. Platinum-lined melting vessels are usually required to prepare glasses with the high optical homogeneity needed for laser applications. However, the liners can generate trace concentrations of microscopic Pt metal particles in the glass. Although very small to begin with, inclusion damage can grow with successive laser shots to several millimeters or even centimeters in size, eventually making the laser glass unusable. Also, large damage spots (>0.3 mm) in the laser glass can seed damage in other optics in the laser chain. The presence of high levels of Pt inclusions typically found, for example, in silicates, borosilicates, and fluorophosphates, make these glasses unsuitable for high-energy laser applications. In contrast, inclusion-free laser glasses can be made using phosphate-based glass compositions melted under highly oxidizing conditions.26–28

Today's laser glass melting systems are designed to both minimize inclusion generation and dissolve any residual inclusions in the glass. The effects of glass composition on platinum solubility have been reported to follow the trend: phosphate >silica-phosphate≫ fluorophosphates >silicate.64 These results are based on solubility measurements using LHG-5 and LHG-8 (phosphates), HAP-3 (silica-phosphate), LHG-10 (fluorophosphate), and LSG-91H (silicate). In a similar study, Hayden et al.65 have examined the effects of the Al2O3 concentration in phosphate glasses on Pt solubility. They chose three commercial phosphate laser glasses (LG-770, LG-760, and APG-1), each having a different Al2O3 content. The effect of alumina was studied because it is a common modifier added to improve thermal–mechanical properties and is often added to HAP phosphate glasses. These researchers report that at higher Al2O3 concentrations the Pt solubility is reduced. The effect of Al2O3 on Pt solubility tends to parallel those reported by Izumitani et al.64 for SiO2 in Hoya HAP-3 glass. This observation agrees with production melting experience in that HAP glasses are generally more difficult than HEHP glasses to manufacture free of Pt inclusions.

Inspection methods have been developed and put into production to scan each piece of laser glass with a high-fluence laser beam to detect the presence of inclusions.66

If Pt inclusions are eliminated, the next limiting factor is the damage resistance of the polished glass surface.67 Consequently, optical material researchers have focused attention on further improving the quality of polished glass surfaces to increase the laser damage resistance. Work to date suggests that the onset of laser-induced surface damage is associated with residual subsurface defects and nanoscale contamination left from the polishing process.67–69 The use of improved polishing methods in combination with postpolishing treatment of the surface to reduce or remove contamination and defects has shown promise for improving surface laser damage resistance for not only phosphate-based glasses but also fused silica.70,71 Fused silica is widely used for passive optic elements (such as lenses, windows and beam splitters) for high-power lasers.

Laser Glass Compositions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Glass Lasers: An Overview
  5. Current High-Power Glass Lasers
  6. Laser Glass Properties Important for High-Power Laser Applications
  7. Laser Glass Compositions
  8. Laser Glass Manufacturing
  9. Glasses to Meet Future HEHP, HAP, and PW Laser Needs
  10. Acknowledgments
  11. References

Development of improved laser glass compositions has been an active research field for more than 40 years. Composition studies have spanned a wide range of glass-forming systems (silicates, phosphates, silicophosphates, fluorophosphates, and fluorides), and in many cases, within each system the effects of variations in network modifiers have been studied (see, e.g., Campbell and Suratwala,8 and sources cited therein).

The early laser glass composition work soon lead to phosphates as the glass of choice and today they are used in all high-power glass laser systems around the world. Hundreds of different phosphate glass compositions have been melted in an effort to simultaneously optimize the laser properties while maintaining acceptable optical, thermal–mechanical, and physical–chemical properties. Figure 12 is a ternary composition diagram for the P2O5–(Al2O3, RE2O3)–(MO, M2O) system showing the compositional region for high-power laser glasses most widely used today. The HEHP glasses lie near the meta-phosphate join (O/P=3) and have the approximate molar composition 60P2O5–10Al2O3–30M2O/MO. Nd is added to this base composition at concentrations of about 0.2 mol% (∼5 × 1019 ions/cm3) for laser rods and up to 1–2 mol% (∼2.5–5 × 1020 ions/cm3) for disks and plates.


Figure 12.  Ternary composition diagram for the P2O5–Al2O3/RE2O3–MO/M2O system showing the approximate compositional region (as indicated by the spot) for most commercial HEHP and HAP laser glasses. These glasses have an O/P ratio of about 3/1 (i.e., “metaphosphate”) implying a backbone structure of rings and long chains of PO3 groups. The units of the plot are in mole fraction.

Download figure to PowerPoint

The key properties of the most widely used commercial laser glasses are summarized in Tables III and IV. The reported compositions for a few of the glasses are listed in Table VI; compositions for others remain proprietary. The range in component values account for small compositional variability due to doping and melting methods and to protect certain proprietary information. The reasons these specific modifiers and compositions make good laser glasses have recently been reviewed.8

Table VI.   Representative Compositions of Commercial HEHP and HAP Nd-Doped Phosphate Laser Glasses and One Developmental Glass, Fujimoto et al.78 (Nd-SG)
Component oxide equivalent (mol%)LHG-88LG-7708LG-7508HAP-352Nd-SG78
  • *

    Nd-doping levels vary depending on use.

  • HAP, high average power; HEHP, high-energy/high-power.

O/P (± 0.1)3333.2NA

Like their HEHP counter-parts, HAP glasses have near meta-phosphate compositions; however, the concentration of Al2O3 and/or SiO2 is generally higher. These constituents give superior thermal–mechanical properties to satisfy the heat loading experienced during HAP operations. Much of the compositional work on HAP glasses has focused on maximizing the glass thermal shock resistance (Eq. [15]). The Nd-doping levels tend to be lower in HAP glasses, generally <∼2.5 × 1020 ions/cm3.

The general development approach for HAP glasses has been the following:

  • 1
    improve the intrinsic thermal–physical properties by appropriate compositional changes, or
  • 2
    modify the glass composition to allow postprocessing to improve strength (e.g., ion exchange), or
  • 3
    carry out both.

Researchers at Schott have taken the compositional route and developed two HAP laser glasses: APG-1 developed in the 1980s,72 and a later version, APG-2, jointly developed with LLNL.49 APG-2 has a thermal shock resistance about two times greater than APG-1 (Table IV).

In another compositional study, workers at Hoya Corporation developed the silica-phosphate glass HAP-373,74 having the base composition 60P2O5, 15SiO2,10Al2O, 13Li2O, 2Nd2O3.52 An improved version, HAP-4, was designed to be further strengthened by ion exchange of the Li+ for larger cations (K+, Na+).75 This exchange process generates a compressive residual stress at the surface. The compressive residual stress offsets the thermally induced tensile stress that develops during operation, allowing the glass to tolerate greater thermal gradients.

Composition research at Kigre Inc. led to a suite of Er3+-, Yb3+-, and Nd3+-doped phosphate glasses that could be ion-exchange strengthened. One such glass, Q-89, is a BaO–Al2O3–P2O5 metaphosphate glass-containing Li2O for ion-exchange strengthening.76,77

Perhaps, the most exciting new area of HAP glass development is the work by Fujimoto and colleagues78–80 who report an Nd-doped high-silica glass made by a zeolite route. The glass has small amounts of Al2O3 to enhance Nd3+ incorporation in the structure. The Al3+concentration is low enough that the glass retains the excellent thermal–mechanical properties associated with the high-silica base. The properties of this developmental glass (Nd-SG) are included in the Table IV summary of HAP glasses to illustrate the much improved thermal shock resistance. Note that certain physical properties of the Nd-SG glass have been estimated by us from values for fused silica.

The zeolite route offers one possible method to avoid OH contamination that has plagued earlier attempts to fabricate Nd-doped silica glasses via a sol–gel route.81 Recall from the early discussion (“Emission cross section and quantum yield”) that residual hydroxyl contamination greatly accelerates the rate of nonradiative energy loss from the upper laser level (4F3/2) to OH vibrational modes.57 Later we discuss the potential role that Nd:silica-type glasses could play in future high-energy laser development.

Laser Glass Manufacturing

  1. Top of page
  2. Abstract
  3. Introduction
  4. Glass Lasers: An Overview
  5. Current High-Power Glass Lasers
  6. Laser Glass Properties Important for High-Power Laser Applications
  7. Laser Glass Compositions
  8. Laser Glass Manufacturing
  9. Glasses to Meet Future HEHP, HAP, and PW Laser Needs
  10. Acknowledgments
  11. References

To a large extent, the development of high-energy and high-peak-power laser systems has been made possible by corresponding developments in advanced manufacturing technology. In the case of laser glass, there have been two key manufacturing developments that enabled the current class of HEHP lasers: continuous melting and high-speed optical finishing. These processes are the result of nearly 10 years of manufacturing development performed separately by Schott and Hoya, for glass melting and Zygo Corporation for glass finishing. This work was carried out in partnership with LLNL and funded by the U.S. Department of Energy.

Continuous Melting

Before 2000, laser glass was melted in a two-step discontinuous process effectively producing one melt (i.e., glass slab) at a time and at a rate of only a few melts per week.82 This technology, apart from being costly, was 20-times too slow to meet the demand for mega-joule scale HEHP laser systems. In addition, the quality of the product achieved with the discontinuous method can vary unpredictably from one melt to the next, simply because of small run-to-run variations in processing conditions.

To meet the HEHP laser glass needs, Schott North America (Duryea, PA) and Hoya Corporation USA (Fremont, CA) each developed continuous glass melting processes.26 These two processes, which use somewhat different proprietary melting technologies, are capable of producing between 70 and 300 glass blanks per week with better optical quality and at a significantly lower cost than possible with the prior pot-melting methods.

Although details of the Hoya and Schott melting systems differ, both carry out the same sequential set of processes. In brief, the continuous laser glass melting process converts high-purity, powdered raw materials into one continuously moving strip of high optical-quality laser glass (Fig. 13). The melting process requires several different operations carried out in separate, but interconnected vessels. The first process step is to mix and dry raw materials with minimal contamination. The laser glass specifications require that the raw materials contain only trace amounts (<10 ppm) of the most common transition metal ions and <0.1 wt% of either physically or chemically absorbed water. The raw material is fed into the system where it dissolves into a pool of molten glass and is mixed by the convection currents inside the melter. The melter consists of custom-designed high-purity refractory material and uses proprietary electrical heating systems.


Figure 13.  Schematic representation of the continuous melting process used to prepare HEHP laser glass blanks from which finished laser slabs are fabricated.

Download figure to PowerPoint

All units beyond the melter are lined with platinum metal (99.9%), as are the interconnecting pipes. As mentioned in a previous section, platinum is required to achieve the part-per-million optical homogeneity necessary for laser applications; however, platinum can contaminate the glass with microscopic metallic inclusions that can initiate laser damage sites (small fractures). To overcome this, a unique conditioner unit was developed that bubbles oxidizing gasses (oxygen and chlorine) through the melt with the dual purpose of dissolving platinum inclusions and minimizing residual hydroxyl contamination in the glass. The reduction in OH content dramatically improves the quantum efficiency. The conditioner unit is perhaps the most complex unit in the melting system.

The glass from the conditioner flows to a refiner section where bubbles are removed using proprietary additives in combination with quiescent conditions and high temperatures. From here, the glass enters the homogenizer and is thoroughly mixed to achieve part-per-million levels of compositional uniformity resulting in an optical homogeneity of Δn≲1 × 10−6. The glass then flows through a platinum tube to a mold where it is formed into a continuously moving strip 5–8 cm thick and about 0.5 m wide. The forming process is designed to produce straie-free glass; straie are regions of high refractive index gradient caused by localized chemical or thermal inhomogeneity (or both). The strip slowly passes through a custom-designed annealing oven roughly 30–35 m long where the glass is gradually cooled from >600°C to room temperature. Glass “blanks,” approximately 1 m × 0.5 m in size, are cut from the end of the strip as it exits the production system (Fig. 13). Each blank is fine annealed, inspected, and fabricated into a prefinished plate that is ready for final grinding and polishing (i.e., “finishing”).

HAP glasses (APG-1and 2, HAP-3 and 4, Q89, etc.) are all manufactured using the two-step discontinuous melting process. This is simply because the demand for these glasses is too small to warrant the expense of using continuous melting technology.

High-Speed Optical Finishing

The goal of the finishing process is to produce a precision optic that can perform its design function (in this case, amplification) without impacting the overall laser beam quality. The beam quality can be adversely affected in three ways by the polished glass surface: (1) aberration of the transmitted wavefront, (2) scattering loss, and (3) laser damage. First, to reduce wavefront aberration the polishing process is used to correct for any large-scale length (≳2 cm) refractive index variations in the bulk glass.83 This requires accurate removal of material from specific locations on the glass surface with a removal-depth accuracy of 10–100 nm. Second, to avoid scattering losses, the residual root-mean-square (rms) “roughness” of the polished surface typically must be <0.4 nm (4Å).83 Scattering from the glass surface can degrade the laser performance by: (a) reducing the output energy of the beam, (b) increasing the noise intensity, which in turn, increases the risk of laser damage due to intensity spikes (“hot spots”), and (c) irradiating surrounding mechanical hardware with scattered light and thus producing vapor or particulate blow-off. The blow-off can contaminate the optic surfaces leading to laser damage and a significant reduction in the useable life of the optic.

To achieve precision-polished glass surfaces, the optic is finished in a series of steps using successively finer abrasives (Fig. 14). During the removal process, the interaction of the abrasive with the glass introduces subsurface defects (fractures) that extend below the glass surface. The goal of the subsequent material removal step is to not only reduce the surface roughness of the part but also to remove sufficient material to get below the affected zone left from the prior step. In general, this requires material removal to a depth three to nine times the mean diameter of the abrasive used on the prior step. At the completion of final polishing, the goal is to remove all finishing-induced subsurface defects and any residual nanoscale polishing contamination, thereby achieving the greatest laser damage resistance and lowest scattering loss.


Figure 14.  Optical “finishing” uses a combination of grinding and polishing steps that successively remove material from the glass surface in incrementally smaller amounts. The goal is to remove sufficient material to eliminate all subsurface fractures (defects) produced by the previous step. The subsurface damage is generally <1 μm deep and widely dispersed at completion of the final precision polishing step.

Download figure to PowerPoint

Application of Edge Cladding

Edge claddings are used to suppress parasitic oscillations of amplified spontaneous emission (ASE) during pumping of large pieces of laser glass.58 Edge cladding consists of a refractive-index-matched glass that is doped (usually with Cu2+) to absorb at the peak Nd-fluorescent wavelength. The cladding is bonded to the edges of the laser glass disks or slabs and absorbs the ASE, thereby preventing parasitic oscillations from developing. Adhesively bonded edge claddings have been developed and are now widely used in the United States, Europe, and Japan.38,84 Two-part optical epoxies and polyurethanes are currently the most widely used adhesives. In general, epoxies are used for HEHP applications84 whereas polyurethanes are more commonly used for HAP lasers.38 Polyurethanes have good moisture resistance and are the best choice for the HAP water-cooled slabs. In contrast, the refractive index of optical epoxies can be very precisely modified to match the glass refractive index84 and are a better choice for HEHP laser slabs with multikilojoules of stored energy.

Commercial Suppliers

The laser glass market generally goes through “boom and bust” cycles that track the R&D and construction phases of the large HEHP laser projects across the world. Such cycles make it difficult for private companies to maintain a commercial presence in the face of other pressing technical needs. For >30 years, there have been three companies that have consistently supplied laser glasses for use in high-power lasers outside of China and Russia: Schott, Hoya and Kigre. Until 2007, Schott and Hoya had been the largest suppliers for the meter-scale glass pieces needed by the HEHP community. Hoya exited the laser glass business in 2007 leaving Schott as the only established commercial source for large sizes and quantities of laser glass. Similarly, Zygo Corporation is currently the only optical finishing company that can cut, clad, grind, and polish meter-scale laser glass parts.

Glasses to Meet Future HEHP, HAP, and PW Laser Needs

  1. Top of page
  2. Abstract
  3. Introduction
  4. Glass Lasers: An Overview
  5. Current High-Power Glass Lasers
  6. Laser Glass Properties Important for High-Power Laser Applications
  7. Laser Glass Compositions
  8. Laser Glass Manufacturing
  9. Glasses to Meet Future HEHP, HAP, and PW Laser Needs
  10. Acknowledgments
  11. References

Future HAP and HEHP glass lasers will need to operate at higher repetition rates with the same or greater output energies. Consequently, improving the laser glass thermal shock resistance with minimum impact to other laser, optical and physical properties will become the main objective of future glass development. To help focus the discussion, Table VII compares key thermal–mechanical properties and the thermal shock resistance of today's commercial laser glasses with other common optical/laser materials.

Table VII.   Thermal Mechanical Properties and the Computed Thermal Shock Resistance (Eq. [15]) for Common Commercial Phosphate Laser Glasses and Other Well-Known Laser Optical Materials
MaterialFracture toughnessThermal conductivityPoisson's ratioYoung's modulusThermal expansionThermal shock resistanceImprovement vs LG-770
(MPa m1/2)(W/m K) GPa10−6/K**W/m1/2
  1. HAP, high average power; HEHP, high-energy/high-power.

I. HEHP glasses
II. HAP glasses
III. Silica glasses
 Fused silica0.81.30.1672.70.5521.867.5
IV. Crystals
 Al2O3 (sapphire)2400.284008.516.952.3

The challenge for HAP lasers is more modest and may be met with modifications to the current suite of glasses. The HEHP requirement is more daunting, and as in the past, will necessitate government investment and support; it is clear that HAP and PW lasers will also benefit from that investment.

HEHP Lasers for Fusion Energy Production

Proposed HEHP lasers for use in Inertial Fusion Energy (IFE)85 and Fusion–Fission Hybrids (LIFE)86,87 call for drivers comprised of diode pumped mega-joule scale systems operating at ∼10 Hz (10 MW). Of course, current designs are still conceptual and clearly will evolve as new pump sources, optical materials, and laser architectures are developed. Also, a future demonstration of PW driven fast-ignition44 would greatly change the driver requirements.

Recall from Table I that current HEHP laser systems operate every few hours (∼10−4 Hz). Therefore, achieving the ∼105 increase in repetition rate needed for IFE power production will be a major optical material challenge. In particular, thermal shock resistance will be a key driver in the choice of a gain medium for the main amplifiers of fusion energy lasers.

Nd-glass is a current leading candidate for the gain media for IFE/LIFE lasers and a successful laser glass will require both HAP and HEHP properties. The best presently available commercial glasses are Schott's APG series (Table IV). However, surface imperfections and/or laser-induced damage in the glass could seed thermal fracture under the power loads at ∼10 Hz operation. Therefore, an inclusion-free gain medium with a much higher Rs (10 ×) and acceptable laser properties would be a great benefit.

Apart from dramatically improved thermal–mechanical properties, the gain medium must be manufactured in large volumes, free of damaging inclusions and with excellent optical quality. Mega-joule lasers require ∼105 L of gain medium (∼300 tons) and production outputs of tons/day to be practical. Arguably, this requirement eliminates single-crystal gain media as an alternative to glass. Currently, the fastest growth rate achieved for large optical single-crystals is solution growth of potassium dihydrogen phosphate (KDP88). Plates of KDP (∼40 cm × 40 cm × 1 cm) cut from large single crystals are used for nonlinear optical applications in today's HEHP laser systems. These crystals require two months to grow to the required ∼500 lb size with an average growth rate of ∼4 kg/day. Although this rate represents a dramatic improvement in single crystal growth, it is ∼1000 times lower than today's continuous optical glass melting rates. Moreover, KDP represents a best case; current high-temperature crystal growth rates (as would be used for laser crystals) are much lower than the rate for KDP (perhaps, 10–100 × in the most optimistic cases). Therefore, growth rate improvements of at least 103–105× would be needed for high-temperature single crystals to be practical.

In contrast, optical glass melting is a mature technology with demonstrated high-production rates. Laser glass melting technology demonstrated tons per day for both LG-770 and LHG-8. This suggests that staying with Nd-glasses, but looking at alternatives with higher thermal shock resistance is a promising approach for early IFE laser development and deployment. This could be achieved by a two-pronged glass approach: one near-term development effort aimed at improving current phosphate glasses and the other a longer term effort to yield a greatly improved alternative. Both approaches are briefly discussed here.

Modified APG Glasses to give Higher Rs. The trend in APG glass development has been toward improving thermal–mechanical properties while accepting some detrimental impact on laser properties, specifically lower cross section and higher saturation fluence.49 These glasses tend to be more difficult to manufacture than today's HEHP glasses because the APG compositions have greater melt viscosities and are more difficult to produce free of platinum particles. Nevertheless, modest improvements in the thermal shock resistance (3–5 ×) appear possible without serious impact on laser properties while maintaining the production technology within reach.

One alternate development approach is to off-set the thermal burden on the laser glass with improved amplifier designs and associated cooling methods. The design of the next generation HEHP lasers is still largely conceptual and one can expect significant design improvements as these projects go forward. It is not unreasonable to expect future design improvements to significantly reduce the thermal shock requirements of the glass.

Nd-Silica Glasses. For decades, it has been recognized that Nd-doped fused silica would make an excellent laser glass because of its high thermal shock resistance (Table VII) coupled with the ability to be manufactured inclusion-free in large sizes with excellent optical homogeneity. The difficulty has been achieving Nd-doping levels ≳1 × 1019/cm3 without clustering; Nd clustering effectively quenches the radiative lifetime as discussed in “Emission cross section and quantum yield” and Fig. 10. Arai et al.89 showed that co-doping SiO2 with Al3+ via CVD inhibited Nd-clustering and thus opened up a new area of research on Nd–Al–SiO2 glasses.

Subsequently, Thomas et al.81 used a Nd–Al–SiO2 sol–gel route to incorporate Nd followed by heating and sintering to densify the sample. Successful removal of residual OH was problematic and the quantum efficiency was low.

The work by Fujimoto and colleagues78–80 at Osaka University has produced, arguably, the most advanced of the Nd–Al–silica glasses to date. These researchers use a zeolite route to incorporate Nd without clustering in an SiO2 matrix with some Al2O3. They showed that the Nd remains in the unclustered state after subsequent high-temperature processing to densify the zeolite preform. Reported properties of this glass (glass code: Nd–SG) are listed in Tables IV and VII for comparison with today's commercial HAP laser glasses. The thermal shock resistance is more than 50 times greater than current HEHP phosphate glasses. Although the authors originally developed, the Nd–SG glass specifically for potential Fusion Energy laser applications79 they also realize the potential for use in HAP lasers.78

The Osaka Nd–Al–SiO2 glass is a remarkable achievement, and in our opinion, represents an excellent starting point rather than end product for future IFE glass development. One reason it may not qualify as a final product is that compared with today's HEHP phosphate glasses the cross section is lower (1.4 × 10−20 vs ∼3.5 × 10−20/cm2) and the associated saturation fluence is higher (2.5 ×). The higher saturation fluence greatly increases the risk of optical damage in the main amplifier beam-path. However, it is likely that with a concerted development effort the composition could be altered to improve these important properties. There is a precedent and template for such an effort; for nearly 30 years, the U.S. Department of Energy funded glass development for Inertial Fusion Research resulting in today's phosphate laser glasses. In particular, systematic compositional studies were carried out in a partnership between the major optical glass companies with LLNL, the University of Rochester LLE, and other laboratories. Measurements of laser, optical, and physical properties were conducted on small samples using standard glass characterization methods. Many of these results are compiled in a set of glass catalogs published by Stokowski et al.90 at LLNL. Clearly, such an approach to developing Nd-SG compositions would be warranted given the success by Fujimoto and coleagues. Parallel efforts in manufacturing development, again similar to what was carried out for phosphates, could also be undertaken. For example, an Nd–Al–SiO2 sol–gel route, as explored earlier by Thomas et al.,81 may be possible with improved processing to remove residual OH.

Higher Repetition Rate HAP Lasers

HAP lasers capable of KW outputs in high-quality ∼20 J single-pulses would offer significantly greater processing rates for materials processing, for example LP. The direction of future HAP glass research would closely follow any efforts made to address improved glasses for HEHP lasers. However, in contrast to HEHP, HAP laser systems may be able to compromise on some degradation in laser properties that would accompany significant improvement in thermal–mechanical properties.

The manufacture of an improved phosphate HAP laser can, in principle, be carried out using today's glass melting technology. However, the manufacturing yields for inclusion-free glass, specifically platinum particle-free, will probably be reduced with the corresponding improvements of thermal–mechanical properties.64

Possible alternatives to phosphate glasses are Nd:Al–SiO2 glasses similar to that being developed by Fujimoto and colleagues discussed above. These authors suggest that a 100 J, 10 Hz laser (1 KW) is within reach91 and list researchers at Shin-Etsu Quartz Products Co. as collaborators indicating a first step toward product commercialization. The viability of this alternative awaits publication of details on glass homogeneity and beam quality data.

Advanced PW Lasers

As discussed earlier, PW lasers fall into two main categories: (i) table-top systems that fit into a large room and (ii) building-size systems that produce high-energies (>1 kJ) as well as PW powers. The latter high-energy systems included the PW lasers associated with major HEHP facilities in the United States, Japan, France, England, and China as reviewed in “HEHP lasers for inertial fusion research.” In some aspects, the missions of the table-top and high-energy systems differ, although there is significant overlap.

The high-energy (∼1 kJ) lasers operate at longer pulse lengths making today's HEHP laser glasses acceptable for the time being. In fact, most high-energy PW systems use amplifiers identical to those of the HEHP main laser system. Thus, any new glasses for these systems will obviously benefit from corresponding future HEHP laser development as discuss above.

Table-top PWs must operate at extremely short pulse lengths to achieve PW powers using only modest output energies. The push to exawatt powers (1018 W) with the associated sub 100-femtosecond pulse lengths is a challenge with today's HEHP glasses, as discussed earlier. New glasses with much broader band widths are needed to handle shorter pulse lengths. Nd–SiO2 glasses are possible options, not because of the much greater thermal shock resistance, but due to the greater bandwidth (∼50 nm78). Although a single glass would offer logistical advantages, the most promising approach, and one in use today,40 is to combine different glasses in series to achieve a larger effective emission bandwidth than offered by a single glass. The combination of current phosphate and silicate laser glasses with a very broad-band Nd-SiO2 glass could significantly extend the effective operating bandwidth for future PW needs.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Glass Lasers: An Overview
  5. Current High-Power Glass Lasers
  6. Laser Glass Properties Important for High-Power Laser Applications
  7. Laser Glass Compositions
  8. Laser Glass Manufacturing
  9. Glasses to Meet Future HEHP, HAP, and PW Laser Needs
  10. Acknowledgments
  11. References

The authors gratefully acknowledge the helpful discussions with scientific and engineering staffs at Schott, LLNL, and Metal Improvement Corporation. The valuable information provided by Dr. Yasushi Fujimoto on his work with Nd-silica is greatly appreciated.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Glass Lasers: An Overview
  5. Current High-Power Glass Lasers
  6. Laser Glass Properties Important for High-Power Laser Applications
  7. Laser Glass Compositions
  8. Laser Glass Manufacturing
  9. Glasses to Meet Future HEHP, HAP, and PW Laser Needs
  10. Acknowledgments
  11. References
  • 1
    E. Snitzer, “Optical Maser Action of Nd3+ in a Barium Crown Glass,” Phys. Rev. Lett., 7, 444446 (1961).
  • 2
    E. Snitzer. “Interview of Dr. Elias Snitzer by Joan Bromberg on August 6, 1984,” Niels Bohr Library & Archives, American Institute of Physics, College Park, MD. Available at
  • 3
    W. Koechner, Solid-State Laser Engineering, 5th edition, Springer-Verlag, New York, 1999.
  • 4
    B. E. A. Saleh and M. C. Teich, Fundamentals of Photonics. John Wiley and Sons Inc., New York, 1991.
  • 5
    A. Siegman, Lasers. University Science Books, Mill Valley, CA, 1986.
  • 6
    W. Koechner and M. Bass, Solid-State Lasers: A Graduate Text. Springer-Verlag, New York, 2003.
  • 7
    R. C. Powell, Physics of Solid-State Laser Materials. Chapter 9. Springer-Verlag, New York, 1998.
  • 8
    J. H. Campbell and T. I. Suratwala, “Nd-Doped Phosphate Glasses for High Energy/High Power Applications,” J. Non-Cryst. Solids, 263 and 264, 318334 (2000).
  • 9
    J. D. Lindl, Inertial Confinement Fusion: The Quest for Ignition and Energy Gain Using Indirect Drive. Springer-Verlag, New York, 1998.
  • 10
    J. D. Lindl et al., “The Physics Basis for Ignition Using Indirect-Drive Targets on the National Ignition Facility,” Phys. Plasmas, 11, 339491 (2004).
  • 11
    H. Azechi, et al., “The FIREX Program on the Way to Inertial Fusion Energy,” The Fifth International Conference on Inertial Fusion Sciences and Applications (IFSA2007), IOP Publishing, J. Phys.: Conf. Ser., 112, 2008, 012002, doi: DOI: 10.1088/1742-6596/112/1/012002; J. Phys.: Conf. Ser., 112, 012002. Available at
  • 12
    Z. Wanguo, et al., “Status of the SG-III Solid-state Laser Facility,” The Fifth International Conference on Inertial Fusion Sciences and Applications (IFSA2007), IOP Publishing, J. Phys.: Conf. Ser., 112, 2008, 032009, doi: DOI: 10.1088/1742-6596/112/3/032009.
  • 13
    X. T. He and W. Y. Zhang, “Inertial Fusion Research in China,” Eur. Phys. J. D, 44, 227231 (2007).
  • 14
    S. V. Grigorovich et al., “Prototype Disc Amplifier for ISKRA-6 Facility,” J. Phys. IV France, 133, 649652 (2006).
  • 15
    J. Ebrardt and J. M. Chaput, “LMJ on its Way to Fusion,” The Sixth International Conference on Inertial Fusion Sciences and Applications, J. Phys.: Conf. Ser.; 244, 2010, 032017, doi: DOI: 10.1088/1742-6596/244/3/032017.
  • 16
    T. R. Boehly et al., “Initial Performance Results of the Omega Laser System,” Opt. Commun., 133, 495506 (1997).
  • 17
    R. L. McCrory et al., “Progress in Direct-Drive Inertial Confinement Fusion Research,” Phys. Plasmas, 15 [5] 055503 (2008).
  • 18
    B. M. Van Wonterghem et al., “Performance of a Prototype for a Large-Aperture Multipass Nd: Glass Laser for Inertial Confinement Fusion,” Appl. Opt., 36 [21] 49324953 (1997).
  • 19
    P. K. Rambo et al., “Z-Beamlet: A Multi-Kilojoule, Terawatt-Class Laser System,” Appl. Opt., 44 [12] 24212430 (2005).
  • 20
    C. A. Haynam et al., “National Ignition Facility Performance Status,” Appl. Opt., 46, 32763303 (2007).
  • 21
    E. I. Moses, R. N. Boyd, B. A. Remington, C. J. Keane, and R. Al-Ayat, “The National Ignition Facility: Ushering in a New Age for High Energy Density Science,” Phys. Plasmas, 16, 041006, 13pp (2009).
  • 22
    Corporate: Schott North America Inc. Advanced Optics. Schott North America Inc., Duryea, PA. Available at
  • 23
    Hoya Corporation USA, Optical Glass Specifications, Hoya Corporation USA, Optics Division, Santa Clara, CA, 2010. Available at
  • 24
    Kigre Inc., Kigre Inc., Hilton Head, SC. Available at
  • 25
    J. -P. Tang et al., “Properties of Phosphate Laser Glass by Continuous Melting,” ACTA Photonica Sin., 37, 248251 (2008).
  • 26
    J. H. Campbell et al., “Continuous Melting of Phosphate Laser Glasses,” J. Non-Cryst. Solids, 263 and 264, 342357 (2000).
  • 27
    J. H. Campbell, et al., “Elimination of Platinum Inclusions in Phosphate Laser Glasses,” Lawrence Livermore National Laboratory Report UCRL-JC-53932, 1989.
  • 28
    J. H. Campbell et al., “Effects of Process Gas Environment on Platinum Inclusion Density and Dissolution Rate in Phosphate Laser Glasses,” Glastech. Ber. Glass Sci. Technol., 68 [2] 5969 (1995).
  • 29
    J. H. Campbell et al., “Effects of Melting Conditions on Platinum Inclusion Content in Phosphate Laser Glasses,” Glastech. Ber. Glass Sci. Technol., 68 [1] 1121 (1995).
  • 30
    C. B. Thorsness et al., “Dehydroxylation of Phosphate Laser Glass,” SPIE, 4102, 175194 (2000).
  • 31
    T. Suratwala, et al., “Technical Advances in the Continuous Melting of Phosphate Laser Glass,” Inertial Fusion Sciences and Applications, ed. K.Tanaka, D.Meyerhofer, and J.Meyert-ter-Vehn. Elsevier, Amsterdam, 540543, 2002.
  • 32
    E. I. Moses et al., “The National Ignition Facility: The World's Largest Optics and Laser System,” SPIE, 5001, 115 (2003).
  • 33
    G. H. Miller et al., “The National Ignition Facility,” SPIE, 5341, 84101 (2004).
  • 34
    K. Ding and L. Ye, Laser Shock Processing: Performance and Simulation. CRC Press LLC, Boca Raton, FL, 2006.
  • 35
    A. Kruusing, Handbook of Liquids-Assisted Laser Processing. Elsevier, Oxford, U.K.
  • 36
    Metal Improvement Company Inc., Laser Peening Division, Livermore, CA. Available at (accessed December 2010).
  • 37
    LSP Technologies., Dublin, OH. Available at (accessed December 2010).
  • 38
    C. B. Dane et al., “Design and Operation of a 150 W Near Diffraction-Limited Laser Amplifier with SBS Wavefront Correction,” IEEE J. Quantum Electron., 31, 148163 (1995).
  • 39
    C. B. Dane, et al., “High Throughput Laser Peening of Metals using a High-Average-Power Nd:Glass Laser System,”International Forum on Advanced High Power Lasers and Applications, Osaka, Japan, November 1–5, 1999; LLNL Report UCRL-JC-134266, November 1999, 1–13.
  • 40
    M. D. Perry and G. Mourou, “Terawatt to Petawatt Sub-Picosecond Lasers,” Science, 264, 917924 (1994).
  • 41
    M. D. Perry et al., “Petawatt Laser Pulses,” Opt. Lett., 24, 160162 (1999).
  • 42
    G. A. Mourou et al., “Ultrahigh-Intensity Lasers: Physics of the Extreme on a Tabletop,” Phys. Today, 51, 2228 (1998).
  • 43
    D. Umstadter, “Review of Physics and Applications of Relativistic Plasmas Driven by Ultra-Intense Lasers,” Phys. Plasmas, 8, 17741785 (2001).
  • 44
    M. Tabak et al., “Ignition and High Gain with Ultra-Powerful Lasers,” Phys. Plasmas, 1, 16261634 (1994).
  • 45
    International Committee on Ultra-High Intensity Lasers, International Committee on Ultra-High Intensity Lasers. Available at
  • 46
    D. Strickland and G. Mourou, “Compression of Amplified Chirped Optical Pulses,” Opt. Commun., 56, 219221 (1985).
  • 47
    H. T. Nguyen, et al., “Gratings for High-Energy Petawatt Lasers,”SPIE Boulder Damage Symposium XXXVII, Boulder, CO, September 19–23, 2005, LLNL Report UCRL-CONF-217007, November 11, 2005.
  • 48
    G. R. Hays et al., “Broad-Spectrum Neodymium-Doped Laser Glasses for High- Energy Chirped-Pulse Amplification,” Appl. Opt., 46, 48134819 (2007).
  • 49
    S. A. Payne et al., “Laser Properties of a New Average-Power Nd-Doped Phosphate Glass,” Appl. Phys. B (Lasers Opt.), B61, 257266 (1995).
  • 50
    S. A. Payne et al., “Spectroscopic Properties of Nd3+ Dopant Ions in Phosphate Laser Glass,” Ceram. Trans.: Solid State Opt. Mater., 28, 253260 (1992).
  • 51
    J. S. Hayden et al.., “Effect of Composition on the Thermal, Mechanical, and Optical Properties of Phosphate Laser Glasses,” High-Power Solid State Lasers Appl., SPIE, 1277, 121139 (1990).
  • 52
    H. Toratani, “Properties of Laser Glasses,” Ph.D. Thesis, Kyoto University, Japan, 1989, 1–187.
  • 53
    B. R. Judd, “Optical Absorption Intensities of Rare-Earth Ions,” Phys. Rev., 127, 750761 (1962).
  • 54
    G. S. Ofelt, “Intensities of Crystal Spectra of Rare-Earth Ions,” J. Chem. Phys., 37, 511519 (1962).
  • 55
    W. Krupke, “Induced Emission Cross-Sections in Neodymium Laser Glasses,” IEEE J. Quantum Electron., QE-10, 450457 (1974).
  • 56
    P. R. Ehrmann and J. H. Campbell, “Nonradiative Energy Losses and Radiation Trapping in Neodymium-Doped Phosphate Laser Glasses,” J. Am. Ceram. Soc., 85, 106169 (2002).
  • 57
    P. R. Ehrmann et al., “Neodymium Fluorescence Quenching by Hydroxyl Groups in Phosphate Laser Glasses,” J. Non-Cryst. Solids, 349, 105114 (2004).
  • 58
    D. C. Brown, High-Peak-Power Nd:Glass Laser Systems. Springer-Verlag, Heidelberg, Berlin, 1981, p. 45.
  • 59
    J. T. Hunt and D. R. Speck, “Present and Future Performance of the Nova Laser System,” Opt. Eng., 28, 461468 (1989).
  • 60
    J. T. Hunt et al., “Hot Images from Obscurations,” Appl. Opt., 32, 59735982 (1993).
  • 61
    N. L. Boling et al., “Empirical Relationships for Predicting Non-Linear Refractive-Index Changes in Optical Solids,” IEEE J. Quantum Electron., QE-14, 601608 (1978).
  • 62
    J. Marion, “Appropriate Use of the Strength Parameter in Solid State Slab Laser Design,” J. Appl. Phys., 60, 69 (1986).
  • 63
    J. E. Marion, “Strengthened Solid-State Laser Materials,” Appl. Phys. Lett., 47 [7] 694696 (1985).
  • 64
    T. Izumitani, et al., “Solubility of Pt in Nd Phosphate Laser Glass,” Laser Induced Damage in Optical Materials: 1987, October 26–28, 1987, NIST Special Publication 756, eds., H.E. Bennett, A.H. Guenther, D. Milam, B.E. Newnan, and M.J. Soileau. U.S. Department of Commerce, National Institute of Standards and Technology, Boulder, CO, 2934, 1988.
  • 65
    Y. T. Hayden et al., “Effect of Phosphate Glass Composition on the Rate of Platinum Dissolution,” Ceram. Trans.: Solid State Optical Mater., 28, 283296 (1992).
  • 66
    C. L. Weinzapfel, et al., “Large Scale Damage Testing in a Production Environment,” Laser Induced Damage in Optical Materials: 1987, October 26–28, 1987, Special Publication 756, eds., H.E. Bennett, A.H. Guenther, D. Milam, B.E. Newnan, and M.J. Soileau. U.S. Department of Commerce, National Institute of Standards and Technology, Boulder, CO, 112112, 1987.
  • 67
    J. H. Campbell, “Damage Resistant Optical glasses for High-Power Lasers: A Continuing Glass Science and Technology Challenge,” First International Workshop on Glass and the Photonics Revolution, May 28–29, 2002, Bad Soden, Germany. Glass Sci. Technol, eds. M.A. Rasmussen, L.D. Pye, and H.A. Schaffer. 75 [C1] 91–108 (2002).
  • 68
    P. E. Miller et al., “Fracture-Induced Sub-Bandgap Absorption as a Precursor to Optical Damage on Fused Silica Surfaces,” Opt. Lett., 35, 27022704 (2010).
  • 69
    T. A. Laurence et al., “Metallic-Like Photoluminescence and Absorption in Fused Silica Surface Flaws,” Appl. Phys. Lett., 94, 151114 (2009).
  • 70
    T. I. Suratwala et al., “HF-Based Etching Process for Improving Laser Damage Resistance of Fused Silica Optical Surfaces,” J. Am. Ceram. Soc., 93, 113 (2010).
  • 71
    T. I. Suratwala et al., “Toward Deterministic Material Removal and Surface Figure During Silica Pad Polishing,” Am. Ceram. Soc., 93, 13261340 (2010).
  • 72
    J. S. Hayden, D. L. Sapak, and H. J. Hoffman, “Advances in Glasses for High Average Power Laser Systems,” Conf. High Power Solid State Lasers, SPIE, 1021, 3641 (1989).
  • 73
    T. Izumitani, et al., “Development of Silicophosphate Glass, HAP-3,” Laser Induced Damage in Optical Materials: 1986, November 3–5, 1986, NIST Special Publication 752, eds., H.E. Bennett, A.H. Guenther, D. Milam, B.E. Newnan, and M.J. Soileau. U.S. Department of Commerce, National Institute of Standards and Technology, Boulder, CO, 13, 1988.
  • 74
    C. Hata, et al., “Chemically-Strengthened Slab Laser Glass with Optical Surface Quality,” Laser Induced Damage Optical Mater: 1986, November 3–5, 1986, NIST Special Publication, 752, eds., H.E. Bennett, A.H. Guenther, D. Milam, B.E. Newnan, and M.J. Soileau. U.S. Department of Commerce, National Institute of Standards and Technology, Boulder, CO, 168175, 1988.
  • 75
    H. C. Lee and H. E. Meissner, “Ion-Exchange Strengthening of High Average Power Phosphate Laser Glass,” SPIE, 1441, 87103 (1991).
  • 76
    S. Jiang et al., “Chemically Strengthened Er3+, Nd3+ Doped Phosphate Laser Glasses,” SPIE, 2379, 1725 (1995).
  • 77
    U. Griebner et al., “Laser Performance of a New Ytterbium Doped Phosphate Laser Glass,” OSA Proc. Adv. Solid State Lasers, 14 (1996).
  • 78
    Y. Fujimoto et al., “Development of Nd-Doped Optical Gain Material Based on Silica Glass with High Thermal Shock Parameter for High-Average-Power Laser,” Jpn. J.Appl. Phys., 44, 17641770 (2005).
  • 79
    Y. Fujimoto and M. Nakatsuka, “A Novel Method for Uniform Dispersion of the Rare Earth Ions in SiO2 Glass Using Zeolite X,” J. Non-Cryst. Solids, 215, 182191 (1997).
  • 80
    T. Sato et al., “Laser Oscillation of Nd-Doped Silica Glass with High Thermal Shock Parameter,” Jpn. J. Appl. Phys., 45, 69366939 (2006).
  • 81
    I. M. Thomas et al., “Optical Properties and Laser Demonstrations of Nd-Doped Sol-Gel Silica Glasses,” J. Non-Cryst. Solids, 151, 183194 (1992).
  • 82
    A. J. Marker, “Optical Glass Technology,”SPIE Proceedings, Geometrical Optics; 531, 2–10 (1985).
  • 83
    J. H. Campbell et al., “NIF Optical Materials and Fabrication Technologies: An Overview,” SPIE, 5341, 85101 (2004).
  • 84
    J. H. Campbell, et al., “Development of Composite Polymer-Glass Edge Claddings for NOVA Laser Disks,” Damage in Laser Materials: 1986, November 3–5, 1986, NIST (U.S.) Special Publication, 752, eds., H. E.Bennett, A. H.Guenther, D.Milam, and B. E.Newnam. U.S. Department of Commerce, National Institute of Standards and Technology, Boulder, CO, 1941, 1987.
  • 85
    M. Yamanaka, et al., “Laser-Diode Pumped 10 J × 10 Hz Nd:Glass Slab Laser for Inertial Fusion Energy,” Inertial Fusion Sciences and Applications 99, eds., C.Labaune, W. J.Hogan, and K. A.Tanaka. Elsevier, Paris, 644649, 2000.
  • 86
    E. I. Moses et al., “A Sustainable Nuclear Fuel Cycle Based on Laser Inertial Fusion Energy,” Fusion Sci. Technol., 56, 547565 (2009).
  • 87
    J. Caird et al., “Nd:Glass Laser Design for Laser ICF Fission Energy (LIFE),” Fusion Sci. Technol., 56, 607617 (2009).
  • 88
    N. Zaitseva and L. Carmen, “Rapid Growth of KDP-Type Crystals,” Prog. Cryst. Growth Charact. Mater., 43, 1118 (2001).
  • 89
    K. Arai et al., “Aluminum or Phosphorus Co-Doping Effects on the Fluorescence and Structural Properties of Neodymium-Doped Silica Glass,” J. Appl. Phys., 59, 3430 (1986).
  • 90
    S. E. Stokowski, et al., “Nd-Doped Laser Glass Spectroscopic and Physical Properties,” Lawrence Livermore National Laboratory Report M-095, Rev. 2, Vol. 1 and 2, 1981.
  • 91
    T. Sato et al., “40J-Class Laser Oscillation of Nd-Silica Glass with High Thermal Shock Parameter,” Appl. Phys. Lett., 90, 221108 (2007).